The techniques used in judging the valuations of companies and projecting their prices in the future are broadly divided into two categories - technical and fundamental analysis. Technical analysis studies the momentum with which prices are moving and attempts to project how far they will continue or if they are about to reverse, attempting to use these movements for profit. As this requires only the price history of the stocks and is easily visualised on charts, it is wildly popular, although sometimes ambiguous.

Fundamental analysis attempts to remove this ambiguity by studying the financial and business details of the company and coming up with a fair value of its stock. Since the price of the stock should eventually align with the reality of its financial fundamentals, people with an idea of its real value can buy up stocks they believe are undervalued. 

Among the two methodologies, fundamental analysis is considered the more logical and hence preferred over technical analysis. However, fundamental analysis has a major problem of accessibility. Online sources usually list only a couple of the most popular metrics used in fundamental analysis. It is also hard to find fundamental data going back more than a couple of years for modeling and evaluation. And finally, probably the biggest problem for anyone other than the most serious investor is a lack of understanding about how to make sense of it all.

The Joel Greenblatt Portfolio

Joel Greenblatt is an American hedge fund manager, academic and author who has written a number of books to educate the general public about investment opportunities. Recognizing the aforementioned difficulties that people face in using fundamental analysis, he wrote The Little Book That Beats The Market to teach people a simple way that they could use fundamental analysis in their investing in plain language.

In the book, he explains that it makes basic economic sense to invest in companies that are able to generate higher profits relative to other companies, but only buying them when they are available at a bargain. The ability of a company to generate profits is measured by its return on assets(ROA) which is the ratio of the company's profits to its assets. It measures how efficient the company is at generating profits from its resources. This prevents the analysis from being weighed by the size of a company. A large company will have to generate more profits to compare against a small company with fewer assets. The price to earnings(P/E) ratio, calculated by dividing the current price of the stock by the earnings per share(EPS) of the company, is used to decide if the price is a bargain. Of two companies with the same EPS, the one available at a lower price is a bargain because you buy a share that is generating the same earnings for a lower price.

Grennblatt explains that his method involves assigning ranks to companies in increasing order of price to earnings and a second set of ranks in decreasing order of return on assets. These ranks are then added and the growth prospects of a company decrease as the sum of its ranks increases i.e. the price of a share is more likely to increase if it has a lower rank. This achieves his objective of finding companies that are good at earning profits and also available at a bargain. Since both P/E and ROA are readily available online, it should be easy to sort and rank companies and implement this strategy. The rest of the book discusses a study the author's hedge fund conducted using stock market data for 17 years and the surprising results the simple method described above gave, providing annualised returns 2.5 times those of the benchmark index.

This Project

Greenblatt is so confident in his model that at the end of the book, he advises readers to simply go to his website magicformulainvesting.com and build a portfolio from a list of stocks generated by the screener there. The portfolio is to be held for one year before being sold off completely. This process is to be repeated every couple of months for a number of years since the Greenblatt method is a long-term investment strategy.

Although the author asserts that the model is based on sound economic principles and should work in other markets, investors outside USA are left in uncertainty since studies similar to the one conducted by the author are unavailable for other markets. This project applies the model on the Indian stock market to provide another point of reference for people curious about the method's effectiveness.

The study involved a significant amount of data collection, specifically, balance sheets and profit & loss statements of more than 1500 companies as well as their price history over a period of 11 years(in all, over 6000 files and 4.5 million rows of data). This data then needed to be extensively processed and modelled, including several steps to account for stock splits, bonuses, rights issues etc., implementing the portfolio model and calculating annualised returns. For this reason, the details of data collection, modeling and analysis are discussed here while the results of the study are discussed below.

Results

A Power BI dashboard was created that selects a portfolio for the month and portfolio size provided by the user and then plots the value of the portfolio as a percentage of the initial investment over the following year. A second worksheet plots the performance of the Joel Greenblatt portfolio for the entire 11 year period. The portfolio for each month is selected and its value at the time of selling is plotted as a percent of the initial investment against the month in which the portfolio was bought to create a consolidated chart of the overall performance of the Joel Greenblatt methodology. Both these charts also include the performance of the benchmark index NIFTY 50 for comparison. The visualisations(embedded below) can also be accessed here.

The month and portfolio size parameters can be changed in the dashboard below to find the components  of the portfolio as well as its performance over the holding period.

Discussion

The characteristics of the model can be observed from the above dashboard. The portfolio outperforms the benchmark index by a wide margin for the majority of time. The following dashboard presents a consolidated picture of the portfolio's performance over the entire 11 year period. Furthermore, the behaviour of the portfolio seen in this graph closely matches the author's description of his own study's results, suggesting that the results from the original study weren't a one-off outcome of the massive bull run in American markets in the late 1990s, but in fact a result of the methodology itself.

The annualised returns from the portfolio are anywhere between 50 to 75% greater than the returns from the benchmark index depending on portfolio size. This significant difference in returns indicates that the Greenblatt methodology definitely merits closer attention for integrating into the overall investment strategy. 

An important consideration regarding returns is that the expected returns from the methodology should be slightly higher than the results presented here. The strategy involves eliminating stocks with P/E less than 5, as such abnormally low values often indicate unusual circumstances for the company. A filter was therefore used to eliminate all stocks whose P/E was below 5. Some stocks with P/E just above 5 were thus included in the analysis and because of extremely low P/E values, were assigned very low ranks(i.e. included in portfolios). Because there is logically no qualitative difference between companies with P/E just below and just above 5, a better strategy would also eliminate the top 10-20 stocks from the ranks list before selecting the portfolio. This was also confirmed in testing with a portfolio consisting of stocks with ranks between 20 and 40 generating returns double that of the benchmark index.